ai.h2o.targetencoding

Class TargetEncoderModel

java.lang.Object

water.Iced<T>

water.Keyed<T>

water.Lockable<M>

hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

ai.h2o.targetencoding.TargetEncoderModel

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, water.api.StreamWriter, water.Freezable<TargetEncoderModel>

public class TargetEncoderModel
extends hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	TargetEncoderModel.TargetEncoderOutput
static class	TargetEncoderModel.TargetEncoderParameters

Nested classes/interfaces inherited from class hex.Model

hex.Model.BigScore, hex.Model.BigScoreChunkPredict, hex.Model.BigScorePredict, hex.Model.Contributions, hex.Model.DeepFeatures, hex.Model.ExemplarMembers, hex.Model.FeatureFrequencies, hex.Model.GetMostImportantFeatures, hex.Model.GetNTrees, hex.Model.GLRMArchetypes, hex.Model.GridSortBy, hex.Model.InteractionBuilder, hex.Model.InteractionPair, hex.Model.InteractionSpec, hex.Model.JavaModelStreamWriter, hex.Model.LeafNodeAssignment, hex.Model.Output, hex.Model.Parameters, hex.Model.StagedPredictions

Field Summary

Fields

Modifier and Type	Field and Description
protected static java.lang.String	ALGO_NAME

Fields inherited from class hex.Model

_dist, _output, _parms, _toDelete, _warnings, _warningsP, scoringInfo

Fields inherited from class water.Lockable

_lockers

Fields inherited from class water.Keyed

_key

Constructor Summary

Constructors

Constructor and Description
TargetEncoderModel(water.Key<TargetEncoderModel> selfKey, TargetEncoderModel.TargetEncoderParameters parms, TargetEncoderModel.TargetEncoderOutput output, TargetEncoder tec)

Method Summary

Modifier and Type	Method and Description
TargetEncoderMojoWriter	getMojo()
hex.ModelMetrics.MetricBuilder	makeMetricBuilder(java.lang.String[] domain)
protected water.Futures	remove_impl(water.Futures fs, boolean cascade)
water.fvec.Frame	score(water.fvec.Frame fr, java.lang.String destination_key, water.Job j, boolean computeMetrics, water.udf.CFuncRef customMetricFunc)
protected double[]	score0(double[] data, double[] preds)
water.fvec.Frame	transform(water.fvec.Frame data, byte strategy, boolean useBlending, BlendingParams blendingParams, long seed) Transform with default noise of 0.01
water.fvec.Frame	transform(water.fvec.Frame data, byte strategy, double noiseLevel, boolean useBlending, BlendingParams blendingParams, long seed) Transform with noise

Methods inherited from class hex.Model

adaptTestForTrain, adaptTestForTrain, addMetrics, addModelMetrics, addWarning, auc, AUCPR, checksum_impl, classification_error, compareTo, computeDeviances, containsResponse, data, defaultThreshold, defaultThreshold, deleteCrossValidationFoldAssignment, deleteCrossValidationModels, deleteCrossValidationPreds, deviance, deviance, exportBinaryModel, exportMojo, fetchAll, fillScoringInfo, getDefaultGridSortBy, getFeaturesUsedInPrediction, getGenModelEncoding, getPojoInterfaces, getToEigenVec, haveMojo, havePojo, importBinaryModel, isFeatureUsedInPredict, isFeatureUsedInPredict, isSupervised, last_scored, lift_top_group, likelihood, logloss, loss, mae, makeBigScoreTask, makeInteraction, makeInteractions, makeInteractions, makeSchema, makeScoringDomains, makeScoringNames, makeScoringNames, mean_per_class_error, modelDescriptor, mse, needsPostProcess, postProcessPredictions, predictScoreImpl, r2, readAll_impl, resetThreshold, rmsle, score, score, score, score, score, score, score0, score0, score0, score0PostProcessSupervised, scoreMetrics, scoring_history, scoringDomains, setupBigScorePredict, testJavaScoring, testJavaScoring, testJavaScoring, testJavaScoring, toJava, toJava, toJava, toJavaCheckTooBig, toJavaInit, toJavaNCLASSES, toJavaPredictBody, toJavaPROB, toJavaSuper, toJavaTransform, toMojo, toString, uploadBinaryModel, writeAll_impl, writeTo

Methods inherited from class water.Lockable

delete_and_lock, delete_and_lock, delete_and_lock, delete, delete, delete, delete, read_lock, read_lock, read_lock, unlock_all, unlock, unlock, unlock, unlock, update, update, update, write_lock_to_read_lock, write_lock, write_lock, write_lock

Methods inherited from class water.Keyed

checksum, readAll, remove_impl, remove, remove, remove, remove, remove, remove, writeAll

Methods inherited from class water.Iced

asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonString, write, writeExternal, writeJSON

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

ALGO_NAME

protected static final java.lang.String ALGO_NAME

See Also:

Constant Field Values

Constructor Detail

TargetEncoderModel

public TargetEncoderModel(water.Key<TargetEncoderModel> selfKey,
                          TargetEncoderModel.TargetEncoderParameters parms,
                          TargetEncoderModel.TargetEncoderOutput output,
                          TargetEncoder tec)

Method Detail

makeMetricBuilder

public hex.ModelMetrics.MetricBuilder makeMetricBuilder(java.lang.String[] domain)

Specified by:

makeMetricBuilder in class hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

transform

public water.fvec.Frame transform(water.fvec.Frame data,
                                  byte strategy,
                                  double noiseLevel,
                                  boolean useBlending,
                                  BlendingParams blendingParams,
                                  long seed)

Transform with noise

Parameters:

data - Data to transform

strategy - A byte value corresponding to TargetEncoder.DataLeakageHandlingStrategy

noiseLevel - Level of noise applied

useBlending - If false, blending is not used when the TE map is applied.If true, blending with corresponding blending parameters are used.

blendingParams - Parameters for blending. If null, blending parameters from models parameters are loaded. If those are not set, DEFAULT_BLENDING_PARAMS from TargetEncoder class are used.

seed -

Returns:

An instance of Frame with transformed data, registered in DKV.

transform

public water.fvec.Frame transform(water.fvec.Frame data,
                                  byte strategy,
                                  boolean useBlending,
                                  BlendingParams blendingParams,
                                  long seed)

Transform with default noise of 0.01

Parameters:

data - Data to transform

strategy - A byte value corresponding to TargetEncoder.DataLeakageHandlingStrategy

useBlending - If false, blending is not used when the TE map is applied.If true, blending with corresponding blending parameters are used.

blendingParams - Parameters for blending. If null, blending parameters from models parameters are loaded. If those are not set, DEFAULT_BLENDING_PARAMS from TargetEncoder class are used.

seed -

Returns:

An instance of Frame with transformed data, registered in DKV.

score0

protected double[] score0(double[] data,
                          double[] preds)

Specified by:

score0 in class hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

score

public water.fvec.Frame score(water.fvec.Frame fr,
                              java.lang.String destination_key,
                              water.Job j,
                              boolean computeMetrics,
                              water.udf.CFuncRef customMetricFunc)
                       throws java.lang.IllegalArgumentException

Overrides:

score in class hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

Throws:

java.lang.IllegalArgumentException

getMojo

public TargetEncoderMojoWriter getMojo()

Overrides:

getMojo in class hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>

remove_impl

protected water.Futures remove_impl(water.Futures fs,
                                    boolean cascade)

Overrides:

remove_impl in class hex.Model<TargetEncoderModel,TargetEncoderModel.TargetEncoderParameters,TargetEncoderModel.TargetEncoderOutput>